updated

app.py

@@ -3,11 +3,8 @@ import torch
 from transformers import (
     AutoTokenizer, 
     AutoModelForSeq2SeqLM,
-    T5ForConditionalGeneration,
-    T5Tokenizer
 )
 from sentence_transformers import SentenceTransformer, util
-import numpy as np
 from typing import List, Tuple, Dict
 import re
 import difflib
@@ -50,7 +47,6 @@ def load_model(model_name: str, model_path: str):
 
 def chunk_text(text: str, max_sentences: int = 4) -> List[str]:
     """Split text into chunks based on number of sentences"""
-    import re
     sentences = re.split(r'(?<=[.!?]) +', text.strip())
     chunks = [' '.join(sentences[i:i+max_sentences]) for i in range(0, len(sentences), max_sentences)]
     return [chunk for chunk in chunks if chunk.strip()]
@@ -79,9 +75,12 @@ def calculate_max_length(input_text: str, mode: str, base_max_length: int) -> in
 
 def calculate_similarity(text1: str, text2: str) -> float:
     """Calculate cosine similarity between two texts"""
+    if not text1.strip() or not text2.strip():
+        return 0.0 
     embeddings = similarity_model.encode([text1, text2], convert_to_tensor=True)
-    similarity = util.cos_sim(embeddings[0], embeddings[1])
-    return similarity.item()
+    similarity = util.cos_sim(embeddings[0], embeddings[1]).item()
+    similarity = round(similarity*100,2)
+    return similarity
 
 def highlight_differences(original: str, generated: str) -> Tuple[str, str, Dict]:
     """
@@ -387,184 +386,6 @@ def process_text(
         return error_msg, "Error occurred", 0.0, "", "", ""
 
 # Create Gradio interface
-# with gr.Blocks(title="Text Paraphraser & Expander", theme=gr.themes.Soft()) as demo:
-#     gr.Markdown(
-#         """
-#         # 📝 Text Paraphraser & Expander
-#         Transform your text with AI-powered paraphrasing and expansion capabilities.
-#         """
-#     )
-    
-#     with gr.Row():
-#         with gr.Column(scale=1):
-#             mode = gr.Radio(
-#                 choices=["Paraphrase", "Expand"],
-#                 value="Paraphrase",
-#                 label="Mode",
-#                 info="Choose to paraphrase or expand your text"
-#             )
-            
-#             model_dropdown = gr.Dropdown(
-#                 choices=list(PARAPHRASE_MODELS.keys()),
-#                 value=list(PARAPHRASE_MODELS.keys())[0],
-#                 label="Model Selection",
-#                 info="Choose the model for processing"
-#             )
-            
-#             gr.Markdown("### ⚙️ Parameters")
-            
-#             temperature = gr.Slider(
-#                 minimum=0.0,
-#                 maximum=2.0,
-#                 value=0.7,
-#                 step=0.1,
-#                 label="Temperature",
-#                 info="Higher = more creative, Lower = more focused"
-#             )
-            
-#             top_p = gr.Slider(
-#                 minimum=0.1,
-#                 maximum=1.0,
-#                 value=0.9,
-#                 step=0.05,
-#                 label="Top-p (Nucleus Sampling)",
-#                 info="Probability threshold for token selection"
-#             )
-            
-#             max_length = gr.Slider(
-#                 minimum=128,
-#                 maximum=1024,
-#                 value=512,
-#                 step=32,
-#                 label="Max Length (tokens)",
-#                 info="Maximum length of generated text per chunk"
-#             )
-            
-#             num_beams = gr.Slider(
-#                 minimum=1,
-#                 maximum=10,
-#                 value=4,
-#                 step=1,
-#                 label="Number of Beams",
-#                 info="Higher = better quality but slower"
-#             )
-            
-#             max_sentences = gr.Slider(
-#                 minimum=1,
-#                 maximum=10,
-#                 value=4,
-#                 step=1,
-#                 label="Sentences per Chunk",
-#                 info="Number of sentences to process together"
-#             )
-            
-#             target_words = gr.Number(
-#                 value=300,
-#                 label="Target Word Count (Expand mode)",
-#                 info="Approximate number of words for expansion",
-#                 visible=False
-#             )
-    
-#     with gr.Row():
-#         with gr.Column(scale=1):
-#             gr.Markdown("### 📥 Input Text")
-#             input_text = gr.Textbox(
-#                 lines=10,
-#                 placeholder="Enter your text here...",
-#                 label="Original Text",
-#                 show_copy_button=True
-#             )
-        
-#         with gr.Column(scale=1):
-#             gr.Markdown("### 📤 Generated Text")
-#             output_text = gr.Textbox(
-#                 lines=10,
-#                 label="Processed Text",
-#                 show_copy_button=True
-#             )
-    
-#     with gr.Row():
-#         process_btn = gr.Button("🚀 Generate", variant="primary", size="lg")
-#         clear_btn = gr.ClearButton([input_text, output_text], value="🗑️ Clear")
-    
-#     stats_display = gr.Markdown()
-    
-#     similarity_display = gr.Number(
-#         label="Cosine Similarity Score",
-#         precision=4,
-#         interactive=False
-#     )
-    
-#     # Highlighted comparison section
-#     gr.Markdown("---")
-#     gr.Markdown("## 🔍 Visual Comparison - See What Changed")
-    
-#     with gr.Row():
-#         with gr.Column(scale=1):
-#             gr.Markdown("### 📄 Original Text (with changes highlighted)")
-#             highlighted_original = gr.HTML(
-#                 label="Original with Changes",
-#                 show_label=False
-#             )
-        
-#         with gr.Column(scale=1):
-#             gr.Markdown("### ✨ Generated Text (with changes highlighted)")
-#             highlighted_generated = gr.HTML(
-#                 label="Generated with Changes",
-#                 show_label=False
-#             )
-
-#     change_stats = gr.HTML(label="Change Statistics")
-    
-#     # Event handlers
-#     mode.change(
-#         fn=update_model_choices,
-#         inputs=[mode],
-#         outputs=[model_dropdown]
-#     )
-    
-#     mode.change(
-#         fn=update_parameters_visibility,
-#         inputs=[mode],
-#         outputs=[target_words]
-#     )
-    
-#     process_btn.click(
-#         fn=process_text,
-#         inputs=[
-#             input_text,
-#             mode,
-#             model_dropdown,
-#             temperature,
-#             top_p,
-#             max_length,
-#             num_beams,
-#             max_sentences,
-#             target_words
-#         ],
-#         outputs=[
-#             output_text, 
-#             stats_display, 
-#             similarity_display,
-#             highlighted_original,
-#             highlighted_generated,
-#             change_stats
-#         ]
-#     )
-    
-#     gr.Markdown(
-#         """
-#         ---
-#         ### 💡 Tips:
-#         - **Paraphrase Mode**: Rewrites text while preserving meaning
-#         - **Expand Mode**: Adds details and elaboration to make text longer
-#         - **Sentences per Chunk**: Controls how many sentences are processed together (4 recommended)
-#         - Adjust temperature for creativity (0.7-1.0 for paraphrase, 1.0-1.5 for expansion)
-#         - Higher beam count = better quality but slower processing
-#         - Max length is automatically calculated based on input, but can be overridden
-#         - Output chunks are separated by double newlines for readability
-#         """
-#     )
 
 with gr.Blocks(title="Text Paraphraser & Expander", theme=gr.themes.Soft()) as demo:
     gr.Markdown(
@@ -663,13 +484,13 @@ with gr.Blocks(title="Text Paraphraser & Expander", theme=gr.themes.Soft()) as d
     
     with gr.Row():
         process_btn = gr.Button("🚀 Generate", variant="primary", size="lg")
-        clear_btn = gr.ClearButton([input_text, output_text], value="🗑️ Clear")
+        clear_btn = gr.Button("🗑️ Clear",size="lg")
     
     stats_display = gr.Markdown()
     
     similarity_display = gr.Number(
-        label="Cosine Similarity Score",
-        precision=4,
+        label="Content Similarity (%)",
+        precision=2,
         interactive=False
     )
     
@@ -682,17 +503,34 @@ with gr.Blocks(title="Text Paraphraser & Expander", theme=gr.themes.Soft()) as d
             gr.Markdown("### 📄 Original Text (with changes highlighted)")
             highlighted_original = gr.HTML(
                 label="Original with Changes",
-                show_label=False
+                show_label=False,
+                elem_id="highlighted_original"
             )
         
         with gr.Column(scale=1):
             gr.Markdown("### ✨ Generated Text (with changes highlighted)")
             highlighted_generated = gr.HTML(
                 label="Generated with Changes",
-                show_label=False
+                show_label=False,
+                elem_id="highlighted_original"
             )
 
-    change_stats = gr.HTML(label="Change Statistics")
+    change_stats = gr.HTML(label="Change Statistics",elem_id="change_stats")
+    
+    gr.HTML("""
+        <style>
+        #highlighted_original > div {
+            overflow-y: auto;  
+            max-height: 400px;   
+        }
+        #highlighted_original > div:empty {
+            overflow: hidden; 
+        }
+        #change_stats > div:empty {
+            overflow: hidden;
+        }
+        </style>
+        """)
     
     # Event handlers
     mode.change(
@@ -730,6 +568,19 @@ with gr.Blocks(title="Text Paraphraser & Expander", theme=gr.themes.Soft()) as d
         ]
     )
     
+    clear_btn.click(
+        fn=lambda: ("", "", 0.0, "", "", ""),
+        inputs=[],
+        outputs=[
+            input_text,
+            output_text,
+            similarity_display,
+            highlighted_original,
+            highlighted_generated,
+            change_stats
+        ]
+    )
+    
     gr.Markdown(
         """
         ---